The awareness of transfer from design to production has stepped up in the past few years and started to get bigger attention and resource investments in the system development process.
Nowadays, many companies are starting to adopt new working approaches with emphasis on better knowledge sharing and integration between different R&D and production departments in order to improve the production procedures and yield.
So, why is there still a problem, even though all companies want to improve their production processes, increase profit and have the willingness to invest a lot of resources and money?
In reality this problem has a many aspect. In this article we at JOYA TEAM would like to share our experience, and present our development process, which we believe can improve the optical product development while taking into account all aspects related to the transfer from design to production.
The awareness of the production process, and the production engineering involvement should take place in the design process starting from the early product design stages, as early as system requirements definitions. This is right for many products, but especially for optical products that have typically very tight and challenging tolerances and high sensitivities. At the optical development kick off, all system requirements should be examined while taking into account also the aspects related to manufacturability and testability of the final product.
Therefore, we would like to review each requirement for performance parameter according to the following points:
How can I test, measure, or calibrate the parameter?
What is the parameter measurement accuracy that is needed?
Is the parameter measurable or is it used for reference to measure other parameters?
In which stage of the production process the parameter should be tested?
To illustrate the importance of these points let’s review an Augmented Reality (AR) system and examine the system Field of View (FOV) requirement.
Note: FOV term definition and our interpretation can be found in our Blog post
Answering the first question requires a better understanding of the system operational requirements, since FOV can be evaluated by:
Demonstration for PASS/FAIL criteria: this can be performed by naked eye inspection or equipment such as digital camera;
Test: this can be done using different types of equipment, such as digital camera, theodolite and motorized equipment that can measure the exact FOV;
Calibration: the desired FOV can be produced by first measuring the FOV and then adjusting it by electronically scaling (up or down) the system display image.
Of course, each possibility has its own pros and cons and the decision which option is most suitable depends on the accuracy needed in the system operational requirement.
For instance, if we look at AR system that features symbology for navigating directions, one should prefer the FOV Demonstration, since the system’s goal is to place all symbology within the system FOV, and very accurate position is not required.
Very similar AR system with a different use, for instance AR system that uses symbology as guidance, such as civilian pilot landing system, where the symbol accuracy is crucial, FOV should be measured and even calibrated to achieve the desired specific tight tolerances.
Giving numerical examples:
If the system FOV requirement is 20° ±1° and the system design provides FOV production tolerances within this limit, then total FOV measurement or demonstration should be sufficient,
If the system FOV requirement is tighter than FOV production tolerances, such as 20° ±0.1°, then FOV measurement won’t be sufficient to maintain high yield in production and therefore FOV calibration procedure should be done.
Another important point that should be considered is understanding whether the required parameter is measured directly, or is it used for reference to measure other parameters?
In order to understand this, point better let’s take the FOV measurement example and review it in this context.
AR system has design requirement of exit pupil position, or Eye Relief, that is defined as the distance between the system surface closest to the eye and the eye position. Placing the eye (or other equipment) at the required Eye Relief is crucial for the FOV measurement accuracy.
Placing the eye at a distance larger than the required Eye Relief will result in smaller visible FOV even though the system is within specification limits. Hence, it is clear that Eye Relief is a reference system parameter and should be considered when planning the FOV measurement setup.
The final point we would like to consider is the impact of the decision in which stage of the production process the FOV should be tested. This decision is crucial for increasing the production yield, but it would have high impact on the system design and performances trade-off.
For instance, FOV may be measured on sub-system level with some reference components that will not be part of the final complete system, which can simplify system testability and speed up the production process, but at cost of measurement coverage and accuracy.
All these considerations should be made as early at system design as possible, in collaboration between the optical designer with the production engineering specialist, so the right decisions and requirements are defined and the design process incorporates manufacturability, production tolerances and testability considerations throughout the design process. This collaboration will ensure that the optical design is optimized to customer needs and requirements while achieving higher yield and cost-effective product in the production line.
コメント